Date Source Site.ID POC
Length:15976 Length:15976 Min. :60010007 Min. :1.000
Class :character Class :character 1st Qu.:60290014 1st Qu.:1.000
Mode :character Mode :character Median :60590007 Median :1.000
Mean :60549600 Mean :1.581
3rd Qu.:60731002 3rd Qu.:1.000
Max. :61131003 Max. :6.000
Daily.Mean.PM2.5.Concentration UNITS DAILY_AQI_VALUE
Min. : 0.00 Length:15976 Min. : 0.00
1st Qu.: 7.00 Class :character 1st Qu.: 29.00
Median : 12.00 Mode :character Median : 50.00
Mean : 16.12 Mean : 53.68
3rd Qu.: 20.50 3rd Qu.: 69.00
Max. :104.30 Max. :176.00
Site.Name DAILY_OBS_COUNT PERCENT_COMPLETE AQS_PARAMETER_CODE
Length:15976 Min. :1 Min. :100 Min. :88101
Class :character 1st Qu.:1 1st Qu.:100 1st Qu.:88101
Mode :character Median :1 Median :100 Median :88101
Mean :1 Mean :100 Mean :88215
3rd Qu.:1 3rd Qu.:100 3rd Qu.:88502
Max. :1 Max. :100 Max. :88502
AQS_PARAMETER_DESC CBSA_CODE CBSA_NAME STATE_CODE
Length:15976 Min. :12540 Length:15976 Min. :6
Class :character 1st Qu.:23420 Class :character 1st Qu.:6
Mode :character Median :40140 Mode :character Median :6
Mean :33270 Mean :6
3rd Qu.:41740 3rd Qu.:6
Max. :49700 Max. :6
NA's :929
STATE COUNTY_CODE COUNTY SITE_LATITUDE
Length:15976 Min. : 1.00 Length:15976 Min. :32.63
Class :character 1st Qu.: 29.00 Class :character 1st Qu.:34.07
Mode :character Median : 59.00 Mode :character Median :35.36
Mean : 54.78 Mean :36.00
3rd Qu.: 73.00 3rd Qu.:37.77
Max. :113.00 Max. :41.71
SITE_LONGITUDE
Min. :-124.2
1st Qu.:-121.4
Median :-119.1
Mean :-119.4
3rd Qu.:-117.9
Max. :-115.5
Date Source Site.ID POC
Length:57775 Length:57775 Min. :60010007 Min. : 1.000
Class :character Class :character 1st Qu.:60311004 1st Qu.: 1.000
Mode :character Mode :character Median :60631007 Median : 3.000
Mean :60571692 Mean : 2.531
3rd Qu.:60771003 3rd Qu.: 3.000
Max. :61131003 Max. :21.000
Daily.Mean.PM2.5.Concentration UNITS DAILY_AQI_VALUE
Min. : -2.200 Length:57775 Min. : 0.00
1st Qu.: 4.200 Class :character 1st Qu.: 18.00
Median : 7.000 Mode :character Median : 29.00
Mean : 8.574 Mean : 32.95
3rd Qu.: 10.900 3rd Qu.: 45.00
Max. :302.500 Max. :353.00
Site.Name DAILY_OBS_COUNT PERCENT_COMPLETE AQS_PARAMETER_CODE
Length:57775 Min. :1 Min. :100 Min. :88101
Class :character 1st Qu.:1 1st Qu.:100 1st Qu.:88101
Mode :character Median :1 Median :100 Median :88101
Mean :1 Mean :100 Mean :88196
3rd Qu.:1 3rd Qu.:100 3rd Qu.:88101
Max. :1 Max. :100 Max. :88502
AQS_PARAMETER_DESC CBSA_CODE CBSA_NAME STATE_CODE
Length:57775 Min. :12540 Length:57775 Min. :6
Class :character 1st Qu.:31080 Class :character 1st Qu.:6
Mode :character Median :40140 Mode :character Median :6
Mean :35447 Mean :6
3rd Qu.:41860 3rd Qu.:6
Max. :49700 Max. :6
NA's :4761
STATE COUNTY_CODE COUNTY SITE_LATITUDE
Length:57775 Min. : 1.00 Length:57775 Min. :32.58
Class :character 1st Qu.: 31.00 Class :character 1st Qu.:34.14
Mode :character Median : 63.00 Mode :character Median :36.60
Mean : 57.02 Mean :36.37
3rd Qu.: 77.00 3rd Qu.:38.10
Max. :113.00 Max. :41.76
SITE_LONGITUDE
Min. :-124.2
1st Qu.:-121.5
Median :-119.8
Mean :-119.7
3rd Qu.:-118.1
Max. :-115.5
Daily PM2.5
summary(epa02$Daily.Mean.PM2.5.Concentration)
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.00 7.00 12.00 16.12 20.50 104.30
summary(epa22$Daily.Mean.PM2.5.Concentration)
Min. 1st Qu. Median Mean 3rd Qu. Max.
-2.200 4.200 7.000 8.574 10.900 302.500
sum(is.na(epa02$Daily.Mean.PM2.5.Concentration))
[1] 0
sum(is.na(epa22$Daily.Mean.PM2.5.Concentration))
[1] 0
There are a total of 20 variables for each year’s EPA summary. The summary data for Daily.Mean.PM2.5.Concentration shows that there are no missing values for 2002 and 2022, however, there is a neagtive minimum value for year 2022, suggesting that we need to take a closer look at the data for issues.
library(dplyr)
Attaching package: 'dplyr'
The following objects are masked from 'package:stats':
filter, lag
The following objects are masked from 'package:base':
intersect, setdiff, setequal, union
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag() masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(data.table)
Attaching package: 'data.table'
The following objects are masked from 'package:lubridate':
hour, isoweek, mday, minute, month, quarter, second, wday, week,
yday, year
The following object is masked from 'package:purrr':
transpose
The following objects are masked from 'package:dplyr':
between, first, last
There are more stations in 2022 compared to 2002, and these new sites especially arose in population dense areas such as San Francisco, Los Angeles, and San Diego. This is important becuase population dense areas are usually polluted due to transportation, hence monitoring PM2.5 in such areas help environmentalists understand patterns and advocate for environment friendly policies.
4. check for missing pm2.5 value and summarize pattern
sum(is.na(epa_merge$dailyPM2.5))
[1] 0
setorder(epa_merge, dailyPM2.5)epa_merge %>%select(Date, dailyPM2.5, dailyAQI, Site.Name)#data frame hidden becuase it's too large
setorder(epa_merge, -dailyPM2.5)epa_merge %>%select(Date, dailyPM2.5, dailyAQI, Site.Name)#data frame hidden becuase it's too large
There does not appear to be a pattern for the best air quality dates in terms of location, however, the lowest daily PM2.5 values appeared in mid-and late January 2002 and 2022. The dates with abnormally high daily PM2.5 values corresponded to the wild fires dates in late summer in California, 2022. For example, the highest daily PM2.5 concentration of 302.5 ug/m^3 LC was recorded in Yreka on July 31st, 2022, corresponding to the McKinney fire on July 29th, 2022. Following sites also coincided with the other 2022 wild fires such as the Mosquito fire in September and October.
5. Explore the main question of interest at three different spatial levels: state, county, sites in LA
State: Histogram
epa_merge$Year1 <-as.factor(epa_merge$Year)
library(ggplot2)epa_merge$Year1 <-relevel(epa_merge$Year1,'2022')ggplot(epa_merge, aes(x = dailyPM2.5, fill = Year1)) +geom_histogram(bins=100, color='black',alpha=0.5,position ='identity') +labs(title="Distribution of sites by Daily PM2.5 Concentration in 2002 and 2022", x="Daily PM2.5 Concentration", y="Count")+xlim(0,100)
# A tibble: 2 × 7
Year mean median sd min max IQR
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 2002 16.1 12 13.9 0.1 104. 13.5
2 2022 8.63 7 7.84 0.1 302. 6.7
California reported more measurements in 2022 since more sites has been built in the 20 years.
The histogram shows a positive skew of PM2.5 concentration for both years.
According to the summary statistics created, the average daily PM2.5 concentration was higher in 2002 (16.12 ug/m^3 LC) than 2022 (8.63 ug/m^3 LC), as well as a higher median, standard deviation, and IQR. The maximum measurement as mentioned in the previous question, was 302.5 ug/m^3 LC in 2022.
According to the EPA, The National Ambient Air Quality Standards (NAAQS) for PM2.5 is set standards thet a 24-hour PM2.5 concentration of 35 ug/m^3 LC and above to be unhealthy for sensitive groups. Therefore to understand the change in unhealthy air quality in the 20 years, a second histogram is created to show the distribution PM2.5 concentration of 35 ug/m^3 LC and above in California. The graph shows that there are less sites that reported unhealthy air quality days in 2022 compared to 2002. In summary, the overall air quality has improved over the 20 years.
County
ggplot(epa_merge) +geom_point(mapping =aes(x = COUNTY, y = dailyPM2.5, colour =factor(Year))) +scale_color_manual(values=c("slateblue", "skyblue")) +labs(x ="County", y ="Daily. PM2.5 Concentration (ug/m^3 LC)") +theme(axis.text.x =element_text(angle =90, vjust = .5, size =5))
`summarise()` has grouped output by 'COUNTY'. You can override using the
`.groups` argument.
# A tibble: 98 × 8
# Groups: COUNTY [51]
COUNTY Year mean median sd min max IQR
<chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 Alameda 2002 14.3 10 11.4 1.9 61.6 10.2
2 Alameda 2022 8.22 7 4.94 0.4 35.5 5.8
3 Butte 2002 14.8 11.5 11.7 1 88 10.5
4 Butte 2022 6.28 4.5 5.78 0.1 42.8 5.2
5 Calaveras 2002 9.9 8 6.50 2 40 6.25
6 Calaveras 2022 6.05 5 4.09 0.5 25.9 4.18
7 Colusa 2002 11.7 9 10.0 1 57 9.5
8 Colusa 2022 7.61 6.7 4.76 0.6 37 6.1
9 Contra Costa 2002 15.1 9.5 14.5 2 76.7 10.1
10 Contra Costa 2022 8.22 7.2 4.93 0.3 37.3 5.6
# ℹ 88 more rows
The highest daily PM2.5 concentrations appeared in mostly in the counties that were heavily influenced by the wildfires in 2022, for example, Mariposa, Nevada, Placer, Riverside, Siskiyou, and Trinity county.
We can notice that the sites with both colors are sites that existed in both 2002 and 2020, among which, the 2022 mean daily PM2.5 concentration was smaller compared to 2002 mean. New sites that were present in 2022 showed average daily PM2.5 concentration lower than 13. Among the new sites, Compton, Long beach route 710-near road, and North Hollywood had the highest mean daily PM2.5 concentration and most often produced maximum values as well. This is possibly due to the high population, high transportation activity, and geographical reasons. Overall, there is an improvement in air quality in California over the last 20 years.
